End of training

README.md

@@ -19,9 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [deepvk/RuModernBERT-base](https://huggingface.co/deepvk/RuModernBERT-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0249
-- Accuracy: 0.9924
-- F1: 0.9923
+- Loss: 0.0236
+- Accuracy: 0.9933
+- F1: 0.9932
 
 ## Model description
 
@@ -41,20 +41,19 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 900
-- eval_batch_size: 900
+- train_batch_size: 750
+- eval_batch_size: 750
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| 0.0472        | 1.0   | 407  | 0.0277          | 0.9914   | 0.9913 |
-| 0.0255        | 2.0   | 814  | 0.0250          | 0.9923   | 0.9922 |
-| 0.0239        | 3.0   | 1221 | 0.0249          | 0.9924   | 0.9923 |
+| 0.0408        | 1.0   | 514  | 0.0242          | 0.9929   | 0.9929 |
+| 0.0243        | 2.0   | 1028 | 0.0236          | 0.9933   | 0.9932 |
 
 
 ### Framework versions